I spent some time to look at the original ticket and testing out some of the scenarios mentioned in pandas==0.25.0 and master.

It looks like only in the exact reported example would an error appear in 0.25.0:

1. Drop Empty List on Index - Returns Same DataFrame [Passed]
2. Drop Empty List on Non-Unique Index - Returns Same DataFrame [Passed]
3. Drop Empty List on DateTimeIndex - Returns Same DataFrame [Passed]
4. Drop Empty List on Non-Unique DateTimeIndex - "TypeError: 'NoneType' object is not iterable" [Failed]

There was no difference between dropping using an empty list vs. empty datetimeindex. However, I decided to test both because list-like values are supported and although a list is the most simple the example used a datetimeindex.

Like mentioned, this is now fixed on master.

I only wrote a test for [3], the very specific scenario in which the Bug was reported. But let me know if anyone thinks we should add more of these scenarios.

Looks like there are some failures on 'Checks' - 'Running doctests'

FAILED pandas/io/formats/style_render.py::pandas.io.formats.style_render.StylerRenderer.format
FAILED pandas/io/formats/style_render.py::pandas.io.formats.style_render._parse_latex_table_styles

I don't think it's related to my Pull Request but going to take a look to see why it's failing or if there's an Open Issue.

Yeah they are unrelated

thanks @mikephung122